#include <iostream>
#include <fstream>
#include <string>
#include <vector>
#include <locale>
#include <unordered_set>

using namespace std;

unordered_set<string> keywords = {
	"procedure",
	"real",
	"var",
	"string",
	"begin",
	"if",
	"then",
	"else",
	"delete",
	"end",
};

unordered_set<string> delims = {
	",",
	"+",
	":",
	")",
	"(",
	";",
	"[",
	"]",
	"<",
	">",
	"'",

};

enum type
{
	keyword,
	delim,
	delim2,
	ident,
	number,
	str,
	error,
};

struct Lexem {
	Lexem(const string &value, type type, int line)
		: value(value)
		, type(type)
		, line(line)
	{}
	string value;
	type type;
	int line;
};

int main()
{
	vector<Lexem> lexems;
	fstream f("C:\\Users\\Rihards\\Desktop\\compiler\\comp.txt", ios::in);
	fstream p("C:\\Users\\Rihards\\Desktop\\compiler\\runtime_output.txt", ios::out);
	string temp;
	int linecount = 1;
	p << "this is start" << endl;
	while (!f.eof())
	{
		while (isspace(f.peek()))
		{
			if (f.peek() == '\n')
			{
				linecount++; p << "end of line" << endl;
			}
			f.get();


		}

		if (f.eof())
			break;

		if (isalpha(f.peek()))
		{
			do {
				temp += tolower(f.get());
			} while (isalpha(f.peek()));
			type t = ident;

			if (keywords.find(temp) != keywords.end()) {
				t = keyword;
			}

			lexems.push_back(Lexem(temp, t, linecount));
			p << temp << " line:" << linecount << endl;
		}
		else if (isdigit(f.peek()))
		{
			do {
				temp += f.get();
			} while (isdigit(f.peek()));
			lexems.push_back(Lexem(temp, number, linecount));
		}

		else //массив делимторов
		{
			if (f.peek() == ':')
			{
				type t = delim;
				temp += f.get();
				if (f.peek() == '=') {
					temp += f.get();
					t = delim2;
				}
				lexems.push_back(Lexem(temp, t, linecount));
			}
			else if (f.peek() == '<')
			{
				type t = delim;
				temp += f.get();
				if (f.peek() == '>') {
					temp += f.get();
					t = delim2;
				}
				lexems.push_back(Lexem(temp, t, linecount));
			}
			else if (f.peek() == '\'')
			{
				lexems.push_back(Lexem("'", delim, linecount));
				f.get();
				while (f.peek() != '\'')
				{
					temp += f.get();
				}
				lexems.push_back(Lexem(temp, str, linecount));

				lexems.push_back(Lexem("'", delim, linecount));
				f.get();
			}
			else
			{
				type t = error;

				temp += f.get();
				if (delims.find(temp) != delims.end()) {
					t = delim;

				}
				else {
					p << "We found error! Line:" << linecount << " UNKNOWN SIMBOL!!!!" << temp << endl;
				}
				lexems.push_back(Lexem(temp, t, linecount));


			}

		}

		temp.clear();
	}



	for (int i = 0; i < lexems.size(); i++) {
		cout << lexems[i].type << '\t' << lexems[i].value << endl;
	}
	cout << "0=keywords, 1=delimiters, 2=double delimiters, 3=identificator, 4=constants, 5=strings, 6=errors" << endl;
	f.close(); //закрывать файловую переменную. файл нам еще пригодится!

	for (int i = 0; i < lexems.size(); i++) {
		if (lexems[i].type == 3)
			cout << lexems[i].type << '\t' << lexems[i].value << endl;
	}


	cout << "___________________" << endl;
	for (int i = 0; i < lexems.size(); i++) {
		if (lexems[i].type == 4)
			cout << lexems[i].type << '\t' << lexems[i].value << endl;
	}

	cout << "___________________" << endl;
	for (int i = 0; i < lexems.size(); i++) {
		if (lexems[i].type == 5)
			cout << lexems[i].type << '\t' << lexems[i].value << endl;
	}

	cout << "_________UNKNOWN SYMBOLS__________" << endl;
	for (int i = 0; i < lexems.size(); i++) {
		if (lexems[i].type == 6)
			cout << lexems[i].type << '\t' << lexems[i].line << '\t' << lexems[i].value << endl;
	}


	system("PAUSE");
	return 0;
}